Вы находитесь на странице: 1из 11

PERFORMANCE COMMENTARY

Notes on the musical text


The v a r i a n t s marked as ossia were given this label by Chopin or were
added in his hand to pupils copies; variants without this designation are the
result of discrepancies in the texts of authentic versions or an inability to
establish an unambiguous reading of the text.
Minor authentic alternatives (single notes, ornaments, slurs, accents, pedal
indications etc.) that can be regarded as variants are enclosed in round
brackets ( ), whilst editorial additions are written in square brackets [ ].
Pianists who are not interested in editorial questions, and want to base their
performance on a single text, unhampered by variants, are recommended to
use the music printed in the principal staves, including all the markings in
brackets.
Chopins original fingering is indicated in large bold-type numerals,
1 2 3 4 5, in contrast to the editors fingering which is written in small italic
numerals 1 2 3 4 5. Wherever authentic fingering is enclosed in parentheses this means that it was not present in the primary sources but added
by Chopin to his pupils copies.
A general discussion on the interpretation of Chopins works is to be contained in a separate volume: The Introduction to the National Edition, in the
section entitled Problems of Performance.

Bar 93 A different fingering of this bar could be the result of


the original text:
5

p. 17

Bar 113 R.H. Execution of the chord with grace-notes:

Bars 114 and 174 R.H. The melody of the theme should be
emphasized:

Abbreviations: R.H. right hand, L.H. left hand.

114

1. Ballade in G minor, Op. 23


p. 12

Bar 25 L.H. The start of the trill with grace-notes:


. d simultaneously with g 1 in the R.H.

174

p. 20

Bar 166
on the first note is obtained easier by playing
3
with the R.H.:
3
1

Bar 32 R.H. The arpeggio should be started together with the


octave in the L.H.
p. 14

Bars 72-75 In order to avoid an excessive mingling of melodic


notes and retain complete harmony, which would be ensured
by Chopinesque pedalling, the latter can be somewhat modified by a simultaneous application of a "harmonic legato" in
the L.H. (fingers sustain the harmonic notes):

p. 21

Bar 179 R.H.


=
.
R.H. Judging by the alignment of the notes in the autograph
(reproduced in our edition) and the approximate calculation of
rhythmic values, the second half of the bar is best executed
as follows:
3

p. 22

Bar 193 R.H. Different fingering:


4

p. 25

p. 16

Bar 92 On modern pianos this bar is best executed with one


pedal since the mingling of melodic notes e 2 and f 2 is less
jarring than losing the basic notes B and f while changing the
pedal on the third quaver. Pianists with a larger hand span
can use the following technique, which makes it possible to
retain full harmony and a clean rendering of the melody:

Bar 246 L.H. The arpeggio should be executed in an anticipatory manner i.e. g 1 simultaneously with e 4 in the R.H.
Bar 258-259 The marking of the breaking of the octaves first
with grace-notes and then with the help of diagonal lines
indicates a gradual acceleration of the break, parallel to the
poco ritenuto accelerando marking, so that together
with the start of bar 260 it would be possible to imperceptibly
enter into ideally simultaneous and rapid octaves.

Performance Commentary

2. Ballade in F major, Op. 38


p. 26

Bars 19, 21, 41, 93 and 95 R.H.

Bar 14 R.H. Facilitation of the second half of the bar:

p. 33

Bar 118 A sign to execute f


editors.

The first note of the arpeggio (in bar 19 c 1 ) should be struck


simultaneously with the appropriate note in the L.H.
p. 31

Using this facilitation, care should be taken so that the sound


effect is close to the sonority resulting from the execution of
the octaves in the R.H. alone, as foreseen by Chopin: not too
rapidly and with suitable articulation legato for the upper
notes and non legato for the lower ones.

with the L.H. comes from the

Bar 165 L.H. The start of the trill with grace-notes:

Bar 22 R.H. In the opinion of the editors the best execution


of the chord grace-note is:

D simultaneously with the first semiquaver in the R.H.


p. 34

Bars 172 and 176 R.H. Facilitation:


5

p. 37

. Analogously in bar 176.

172

Bars 26 and 28 R.H. The start of the trill in bar 26:

f 1 together with the L.H. chord. Analogously in bar 28.


Bars 29-32 R.H. Execution of trills:

5
2

and

5
2

p. 40
1

179

. Analogously in bar 183.

Bar 197 A pianistically convenient and sonorically expressive


realization of the beginning of the bar:

The arpeggiation of certain chords emulates their execution


in the original division into hands (the inclusion of arpeggios
in brackets is left to the discretion of the performer).

p. 41

4 3

Bars 97-98 A different division of chords between two hands:

p. 35

Bars 179 and 183 L.H. Facilitation for a smaller hand:

1
2

Bar 203 R.H. Simultaneously with the octave in the L.H., the
grace-note should be played, and in case the version without
the grace-note is selected the first note of the arpeggio
should be struck.

p. 42

Bar 136 R.H. The grace-note e


with E in the L.H.
Bar 139 R.H.

should be struck together

Playing the lower note of the

arpeggio together with the first L.H. semiquaver is more in


keeping with Chopins style.
p. 45

Bars 189-190 and 197-198 L.H. Facilitation for smaller hands:


2

189

3. Ballade in A flat major, Op. 47


p. 36

Bars 3 and 39 L.H. The grace-note (or in the variant the first
note of the mordent) should be struck simultaneously with the
R.H. chord.
Bars 9 , 10 and analogously The original pedalling seems to
indicate the fact that Chopin wished to retain the sound of the
fundamental bass note also in the second half of the bars
despite the changing harmonies. On modern pianos the retention of the bass notes without blurring the harmonies can
be attained by rapidly changing the pedal in the middle of the
bar. The bass notes can be also sustained with the use of
a third (sostenuto) pedal.

Analogously in bars 197-198.


Bars 190 and 198 R.H. The grace-note should be struck
simultaneously with the first semiquaver of the second half of
the bar in the L.H.
p. 47

Bars 235 and 236 R.H. The start of the trill in bar 235:

d 2 together with the L.H. sixth. Analogously in bar 236.

Performance Commentary

p. 55

4. Ballade in F minor, Op. 52


p. 48

Bar 128 Taking b into the L.H. is an editorial suggestion.


Bar 134 At the end of the cadenza it is possible to retain the
sonority of the basis of harmony, the A-e fifth, without mingling the notes c 1 - d1 - e1- f 1 with a single pedal:

Bar 1 R.H. The fact that Chopin placed the marking legato
between the R.H. voices can denote only a "harmonic legato"
in this case the retention of the notes c 2 as was marked for
the accompanying seconds and thirds in bars 2 and 4.

54 3

Bars 4, 131 and 132 R.H. The lower note of the arpeggios
should be played simultaneously with the L.H.
p. 49

p. 51

Bars 38-45 On modern pianos the pedal depressed at the


beginning of bars 38 and 42 can be sustained longer (at least
by half a bar), and then a pedal changed on each quaver
should be added.
Bars 61 and 65 R.H. The arpeggios should be started together with an appropriate L.H. note.

p. 58

Bars 74-77 On modern pianos the pedal can be sustained at


least a bar longer.

p. 60

Bars 85 and 93 R.H.


crotchet.
p. 53

p. 54

p. 57

with the R.H. comes from the

Bar 65 A sign to execute g


editors.
p. 52

1 2

, b together with the L.H.

Bar 104 L.H. The depression of the pedal immediately after


the third semiquaver makes it possible for even the smallest
hand to sustain the G fundamental bass note without mingling the melodic semitones.
Bars 112 and 114 L.H. The realization of trills in bar 112:

p. 62

Bar 161 R.H.

Bar 173 R.H. The grace-note f 1 before the arpeggio should


be sounded simultaneously with D in the L.H.
3

Bar 191 sqq.

Bars 223 and 225 The Ignaz Friedmann, Alfred Cortot and
Alfred Casella editions contain a variant of fingering worthy of
recommendation:
5
2

4 5( )
1 (2) 2 1

3
1

223

3
1

5
2

5
2

. Analogously in bar 114.


5

3 1 5

Jan Ekier
Pawe Kamiski

SOURCE COMMENTARY

/ABRIDGED/

Introductory remarks

lifetime. The GE version could be either a revision of this


edition (cf. commentary to bars 45 and 47) or a variant introduced by Chopin (to avoid parallel fifths in upper voices?,
together with L e n t o? vide commentary to bar 1). Long
after the death of Chopin, SaintSans stated that he deduced from Liszts elusive response that Chopin played d 1 .
Nonetheless, apart from such certain sources as A and FE,
twice corrected by Chopin, testimonies of four persons closely connected with Chopin (Marcelina Czartoryska, Friederike
Streicher, Ferdinand Hiller and Adolf Gutmann) speak in favour of the version with e 1 .

The following commentary sets out in an abridged form the principles of editing the musical text of particular works and discusses the most important discrepancies between the authentic sources; furthermore, it draws attention to
unauthentic versions which are most frequently encountered in the collected
editions of Chopins music compiled after his death. A separately published
Source Commentary contains a detailed description of the sources, their filiation, justification of the choice of primary sources, a thorough presentation of
the differences between them and a reproduction of characteristic fragments.

Remark to the second edition


This edition of the Ballades took into consideration sources inaccessible during work on the first edition (PWM, Krakw 1967), predominantly: a fragment
of the editorial autograph of the Ballade in F minor, Op. 52 and copies of
different impressions of the first French editions* of all the Ballades, making it
possible to follow their correction by Chopin. This fact enabled a more certain
establishment of the text and the reduction of the number of variants. In comparison to the first edition of Source Commentaries (PWM, Krakw 1970),
data concerning sources and their filiation have been brought up to date (i.a.
the numeration of the impressions of the first editions has been altered).

p. 12

p. 13

p. 14

Abbreviations: R.H. right hand, L.H. left hand. The sign symbolizes
a connection between sources; it should be read "and ... based on it".

1. Ballade in G minor, Op. 23


S o u r c e s
A
Autograph/fair-copy (private collection, photocopy in the
Chopin Society, Warsaw). A served as the basis for the first
French edition.
FE1 First French edition, M. Schlesinger (M.S. 1928), Paris VII
1836. FE1 was based on A and contains numerous changes
introduced by Chopin during the proof-reading.
FE2 The second impression of FE1, VIII 1836, perfunctorily corrected by Chopin.
FE
= FE1 and FE 2 .
FED Collection of copies of FE belonging to Chopins pupil Camille
Dubois, with annotations by Chopin (Bibliotheque
`
Nationale,
Paris), containing fingerings, performance indications, variants and corrections of printing errors.
GE1 First German edition, Breitkopf & Hartel

(5706), Leipzig VI
1836. GE1 repeated (with errors) the FE1 version, introducing its own revisions. We cannot exclude the possibility that
some of the changes in GE1 come from Chopin.
GE 2, GE 3, GE 4 further impressions of GE1, successively introducing unauthentic changes and supplements.
GE
= GE1, GE 2, GE 3 and GE 4 .
EE
First English edition, Wessel & C o (W & C o 1644), London V
1836. EE was based on FE1 and was not corrected by
Chopin.
E d i t o r i a l P r i n c i p l e s
We have accepted as our basis FE2 as the last authentic source,
compared with A.
p. 11

Bar 1 A (FEEE) has L a r g o as the tempo marking. The


L e n t o in GE could have been added by Chopin to the FE1
copy serving as the basis for GE .
Bar 7 L.H. The top note in A (FEEE) is e 1 , and in GE
d 1 . This ambiguity gave rise to doubts already during Chopins

* The inclusion of this group of sources into editorial work was made possible thanks to
the research and assistance of Dr. Krzysztof Grabowski from Paris.

p. 16

Bar 26-27 R.H. In FE (GE , EE) the tie which in A sustains


the minim d 2 to the next bar has been mistakenly deciphered
as a phrasing slur and joined to the slur over bars 27-28.
Bars 45 and 47 R.H. Before the first quaver in both bars GE
has . This is an arbitrary addition made by the reviser of GE
(cf. commentary to the Ballade in F minor, Op. 52, bars 164
and 165).
Bar 63 R.H. The last quaver in GE is mistakenly d . Probably
the engraver mistook this bar for one of the similar bars 57,
59 or 61-62 (such errors were committed elsewhere in GE : d 1
instead of c 1 on the first note of the second half of bar 100
and e 1 instead of d 1 on the fourth crotchet in the L.H. in bar
114). The version with c in A (FEEE) prepares harmony
in bar 64 in a manner characteristic for Chopin, avoiding the
suspension of the d sound, absent in this harmony.
Bar 99 R.H. On the fifth and sixth crotchet A has the chord
e 1 - a1- c 2 . Chopin removed both e 1 in the proofs of FE1 (GE ,
EE) .
Bars 103-105 L.H. On the last beat A has in each of these
bars a rest instead of the chord e-a-c1 . We give the version
Chopin introduced in the proofs of FE1 (GE ,EE). The majority of the later collected editions give an unauthentic compiled version of those bars: bars 103-104 in the original sonority (with a rest) but bar 105 with an arpeggiated chord of
the final version.

p. 17

p. 18

Bars 119 and 123 R.H. In A there are no chromatic signs


over the mordents in these bars. In the proofs of FE1 Chopin
added
over
in bar 123. The majority of the later collected editions, assuming the possibility of an oversight, also
added
in bar 119. There are no arguments, however, in
favour of an oversight by Chopin. The mordent with the semitone g 2 in bar 119 better corresponds to the scale octave
progression in B-minor and the analogous melodic-tonal situation in bar 121. On the other hand, the whole-tonal
in
bar 123 (g 2 - a 2 ) signalizes modulation changes in that particular bar and following bars.
Bars 126 and 136 A has sempre pi animato in bar 126
and p i vivo in bar 136. We give the reduced probably
by Chopin markings in FE (GE ,EE).
Bars 134-135 R.H. A has the original version:
8

Chopin changed it by correcting FE (GE ,EE).


Bar 137 R.H. Before the fifth quaver A has
Chopin in the proofs of FE1 ( GE ,EE).

, removed by

Source Commentary

p. 19

Bar 145 L.H. The original version of the first half of the bar
in A:

p. 20

. Chopin changed it in FE (GE ,EE).

Bar 171 R.H. A (FE1GE ,EE) has a version which we


give in the variant. In FE2 it was changed into the following:
. The alterations were conducted certainly
upon the request of Chopin but it is doubtful whether it was his
intention to leave d 2 in the last but one crotchet of the quintuplet. The erroneous double engraving of the shifted note is
encountered upon several occasions in the first French editions of Chopins works (e.g. the Scherzo in B minor, Op. 20,
bars 135 and 292, the Polonaise in A, Op. 40 No. 1, second
version, bar 93). Hence in the main text we give a version of
FE2 corrected by Chopin, emended analogously to bar 170.

p. 21

p. 22

Bar 173 R.H. The main text comes from FE (GE ,EE), the
variant from A. The crossing-out in A testify to Chopins indecision as regards the two possibilities; similar variants are
encountered in his other works (e.g. the Ballade in A , Op. 47,
bars 3 and 39).
Bar 193 R.H. GE 3 arbitrarily changes the sixth quaver from
e 2 to d 2 .
R.H. The last quaver in A is mistakenly d 1 . Chopin corrected
it to e 1 in the proofs of FE1 (GE ,EE) .
Bars 194, 196 and 198 R.H. The accents in the form of short
lines, rarely used by Chopin, were written by him into FED.

p. 23

Bar 214 L.H. At the beginning of the bar FE (GE ,EE) has
the a-c1 - g1 triad. A is not very legible at this point but there
are no clear reasons for differentiating this bar and bar 210,
with an indubitable four-note chord (with e 1 ).

FE3

The third impression of FE1 (the same firm and number),


which introduced, above all, numerous and, as a rule, obvious supplementations of accidentals. Chopin most probably
did not participate in its production.
FE
= FE1, FE2 and FE3 .
FED As in the Ballade in G minor, Op. 23.
FES ,FEJ collections of pupils copies of FE with annotations by
Chopin, containing fingering, performance directives, variants,
and corrections of printing errors.
FES collection belonging to Chopins pupil, Jane Stirling
(Bibliotheque
`
Nationale, Paris),
FEJ collection belonging to Chopins sister, Ludwika
Jdrzejewiczowa (Chopin Society, Warsaw).
EE
First English edition, Wessel & C o (W & C o 3182), London X
1840. EE was based on A and was not corrected by Chopin.
GE1 First German edition, Breitkopf & Hrtel (6330), Leipzig X
1840. GE1 was based on GC and was not corrected by Chopin. There are copies of GE1 with different prices on the
covers.
GE 2 Second German edition (the same firm and number), after
1852, with small adjustments of the text of GE1 .
GE
= GE1 and GE 2.
E d i t o r i a l P r i n c i p l e s
We have based our text on A and taken into consideration changes
introduced by Chopin to GC and FE2 as well as his additions to
pupils copies.
p. 26

p. 27

p. 28

p. 25

Bar 229 L.H. The majority of the later collected editions


arbitrarily adds b to the authentic g on the last crotchet.
Bar 259 Diagonal lines indicating the breaking of octaves in
the second half of the bar (cf. Performance Commentary )
were omitted in FE (GE ,EE).

2. Ballade in F major, Op. 38


S o u r c e s
A
Autograph/fair-copy which served as the basis for the first
French edition and then the first English edition (Bibliotheque
`
Nationale, Paris).
GC
Copy, probably by Gutmann, which served as the basis for
the first German edition (Stiftelsen Musikkulturens Frmjande,
Stockholm). In GC the copyist committed numerous errors
and imprecisions. In a probably hurried examination, Chopin
introduced a number of supplementations and corrections.
FE1 First French edition, E. Troupenas (T. 925), Paris X 1840.
FE1 was based on A, contained a considerable number of
errors (i.a. in slurring) and was not corrected by Chopin.
FE2 The second impression of FE1 (the same firm and number),
corrected by Chopin.

Bars 37-38 R.H. Analogously to bars 91-92 some of the later


collected editions arbitrarily add notes e 1 on the last quaver
of bar 37 and at the beginning of bar 38.
Bar 53 R.H. The original beginning of the bar was:
. Already in A (FE,EE) Chopin altered it to:

Bar 223 L.H. At the end of this bar FE (GE , EE) has an
a-d1 fourth. This is probably an error caused by the indistinct
record of this crotchet in A. A comparison with the harmonically similar bar 217 and analogous ones speaks in favour of
the a-c1 - d1 chord.
p. 24

Bars 26-27 R.H. Some of the later collected editions arbitrarily tie g 1 notes over bar-line.

. In GC (GE ) Chopin changed this second


version to the text which we present.
Bar 54 L.H. On the fourth quaver some of the later collected
editions arbitrarily changed the A 1 - A octave to C-c.
p. 30

Bar 93 R.H. A (FE,EE) has no arpeggio which is found in


GC (GE ), probably added by Chopin.
Bar 98-99 R.H. A (FE,EE) has no tie sustaining g 1 . Chopin added it in GC (GE ).
Bars 101,107,126 and 132 R.H. In the second half of the bars
it is difficult to ascertain the sonority of the third in the lower
voice intended by Chopin. A (FE1 ,EE) has the version
presented in our edition. In GC (GE ) Chopin added in bar
126, changing b - d1 to b-d1 . In the proofs of FE2 Chopin
added
in bar 132, changing e 1 - g1 to e 1 - g1 . Below we present a list of thirds (major or minor) in particular sources:
bar
101 107 126 132
1. A
major minor major minor
2 . GC
major minor minor minor
3. FE2
major minor major major.
This table shows that:
none of the sources contains an identical version in all
bars
the presumption that while correcting GC or FE2 Chopin
intended to unify those passages leads to a contradiction
since in GC he altered the major third to a minor one, and in

Source Commentary

FE2 the minor third to a major one


accepting that a differentiation of those bars corresponds
to Chopins intentions, we must recognise version 1 as the
most perfect one. The remaining versions are probably the
result of Chopins errors due to a hurried and fragmentary
proof-reading of those greatly similar passages.
Some of the later collected editions, whose point of departure
is the second version, arbitrarily change the major third in bar
101 to a minor one, and give a reduced, identical version in
all four bars.
Bar 105 R.H. In the second half of the bar some of the later
collected editions arbitrarily add a 1 to the f 1 of the lower
voice.
Bar 106 R.H. The main text comes from FE2 in which Chopin
corrected the incomplete and erroneous version of FE1. The
variant is the A version; it is not clear whether it was Chopins
intention to sustain g 1 (as in EE) or to repeat it (GCGE ).
Bars 110-111 R.H. At the end of bar 110 A (FE1 ,EE) has
octaves g 2 - g 3 and a 2- a 3 . Chopin supplemented them with e 3
notes both in GC (GE mistakenly reads f 3 in the last quaver)
and in the proofs of FE2. This corrected version, in which the
passage from a four-note chord in the middle of the bar to an
octave in bar 111, rendered smooth thanks to three-note
chords, is presented by us as the main version. In the variant, however, we cite the original A version since the false
relation between e 3 in bar 110 and the E octaves in bar 111
is less audible. Some of the later collected editions arbitrarily
add e 3 to the octave at the beginning of bar 111.
Bars 111-112 L.H. Octaves B 1 - B are tied between the bars
in A (GCGE , FE1 ,EE). In the proofs of FE2 Chopin removed the ties, probably in order to achieve rhythmic analogy
with bars 136-137. The ties were subsequently restored in
FE3, but probably not by Chopin himself but by the reviser who
in this impression checked the slurs and ties according to A.
We thus present the FE2 version, undoubtedly corrected by
Chopin, as the main version.
p. 31

Bars 123-124 and 129-130 R.H. The majority of later collected editions joins notes f 1 in bars 123-124 and b 1 in bars
129-130 analogously to bars 98-99 and 104-105. Nothing,
however, indicates that Chopin wanted a strict analogy between bars 99-108 and 124-133.
Bars 125-126 L.H. Sources do not lengthen G to bar 126 (no
note and tie). This is certainly Chopins oversight connected
with the fact that bar 126 begins a new page in A. Cf. analogous bars 101,107 and 132.
Bar 133 R.H. In A (GCGE , FE1 ,EE) the semiquaver is
only the g 1- g 2 octave. In the proofs of FE2 Chopin supplemented it with the note c 2.
Bar 139 R.H. FE overlooks b

p. 32

p. 33

p. 34

p. 35

Bars 169 and 173 R.H. There is no accidental before the


lower note of the first sixth in the second half of the bar in A
as well as in GC and FE, both corrected by Chopin. This
means that Chopin wanted to have d 1 in bar 169 and d 2 in
bar 173. The revisers of EE in bar 169 and GE in bar 173
added
next to those notes, perhaps misled by an unnecessary natural next to b 2 in bar 173.
Bar 173 L.H. In A (GCGE , FE1 ,EE) the second quaver
is the c 1- f 1 fourth. This error was corrected by Chopin in the
proofs of FE2.
Bar 196 Before the notes b , b 2 , b 3 in the second half of the
bar A (GCGE1 , FE1FE2) has no accidentals and thus
the flats from the first half of the bar are binding. The crossing-out of the sign, visible in A and GC, before the lower note
of the L.H. chord (GC shows that this was ) also proves that
Chopin checked the correctness of the script in both manuscripts. The revisers of EE, FE3 and GE 2 added naturals
before these notes (b, b 2 and b 3 ), making this chord similar to
two previous ones (in bars 195-196). The authentic version of
the chord (f-a - b -d) avoids a mechanical repetition of a diminished seventh chord in three parts of the sequence; in this
version, the climax chord in bar 197 is more expressive
thanks to three notes, new in relation to the previous chord.
Bars 202-204 The several versions of the end of the Ballade
testify to Chopins undecidedness. In chronological order they
are:
1. The original version in A

(202 )

2. The later version in A (EE), added by Chopin underneath


the crossed out first version

(202 )

This version was copied in GC where Chopin then modified it;


it is also contained in FE1 (with a mistaken c 1 instead of d 1 in
bar 203).
3. The G C version altered by Chopin

(202 )

in the third chord of the bar.

Bar 148 R.H. The upper note in the first semiquaver in A


(FE,EE) is e 1 (our main text). In GC this note was altered
but it is difficult to say whether the mistakenly written f 1 was
corrected to e 1 by Chopin or the copyist, or, on the contrary,
whether Chopin altered the correctly copied e 1 to f 1 (as understood in GE , our variant). Numerous errors and imprecision of the copyist as well as the absence of an accent over
this note, while the f 1 four semiquavers earlier and later are
accented, speak in favour of the first possibility (e 1 ) .
L.H. FE overlooks ties sustaining the G 1 - G octave.

This version was adopted in GE , correcting the lower note in


bar 203 and adding a mistaken tie joining the grace-note and
minim g .
4. The FE2 version corrected by Chopin

(202 )

Source Commentary

These four versions are divided into two pairs with a different
range of the chords: "open" versions 1 and 3 and "close" versions 2 and 4. It is highly probable that it was precisely the
wavering between the "open" ending, corresponding to the
whole Ballade, and the "close" one, which did not transcend
beyond the atmosphere of the few last bars (from t e m p o
p r i m o ), that was the cause of so many changes in this
fragment.
Among the "open" versions, version 3 is later than version 1
but it contains a mistaken pitch (C 1 instead of E 1 in bar 203)
and three distinct stylistic lapses:
the doubling of the leading-note g - g 1 in bar 203
the absence of the resolution of g to a
a double-octave leap of the bass from e in bar 202 to E 1 in
bar 203.
The following conclusion comes to mind: inasmuch as the
very f a c t of corrections conducted by Chopin in GC is the
expression of his unquestioned intention to pass from a "close"
version to an "open" one, the r e s u l t of those corrections,
which is probably due to the hurried manner of their introduction, cannot be acknowledged as definitive. The "open"
version, which best reflects Chopins intention, is, therefore,
the version 1.
Among the "close" versions, version 4, introduced by Chopin
into the proofs of FE2 in place of version 2, should be regarded as definitive.
Our edition takes into consideration Chopins uncertainty by
presenting two basic versions (1 and 4). For the main text we
have accepted the "open" version (enhanced with a variant
grace-note, characteristic for Chopin, and added in GC) which
corresponds to the Ballade as a whole, since while changing
the "close" version in the proofs of FE2 Chopin could have
felt constrained in introducing in print excessively far reaching
alterations.

E d i t o r i a l P r i n c i p l e s
We have based our text on FE2 as the last authentic source compared with A in order to eliminate the numerous errors and oversights of the engravers. We take into consideration Chopins annotations in four pupils copies.
p. 36

Bars 6-7 L.H. In A a tie sustaining the e is found only in bar


7 which starts the new line. The absence of the tie in bar 6
could be Chopins oversight (this is the way it was understood
by [FC]FEEE). It cannot be completely excluded that it
was the tie in bar 7 wich was introduced mistakenly (it is
absent in GE ); a similar Chopins error see Source Commentary to the Fantaisie op. 49, bars 182-183.
Bar 15 L.H. FE has an erroneous rhythm
mistake was corrected by Chopin in FES and FEJ.

p. 38

Bar 46 L.H. Instead of c 2- e

FE (EE) mistakenly has a 1- c 2.

Bars 47-49 The o s s i a variant was added by Chopin in FEJ .

3. Ballade in A flat major, Op. 47

. This

Bar 21 L.H. The last chord in FE1 is f-a - c1- f 1. Correcting this
error, FE2 mistakenly left only the octave f-f 1.

p. 39

S o u r c e s
A
Autograph/fair-copy (lost, photocopy in the Chopin Society in
Warsaw). Fontana used A to make a copy intended as the
basis for the first French edition. Subsequently, A served as
the basis for the first German edition.
[ FC] Lost copy by Fontana, made upon the basis of A. [FC] served
as the basis for the first French edition. Superficial proof-reading by Chopin is not excluded.
SC
Copy by Saint-Saens
(Bibliotheque
Nationale, Paris), made

`
upon the basis of [FC] compared with the first French edition.
SC makes possible an almost complete reconstruction of
[ FC] .
GE1 First German edition, Breitkopf & Hartel

(6652), Leipzig I
1842. GE1 was based on A and was not corrected by Chopin.
GE 2 Second impression of GE1 (the same firm and number) after
1870, correcting part of the errors in GE1 and introducing
a number of changes (some according to FE, others arbitrarily).
FE1 First French edition, M. Schlesinger (M.S. 3486), Paris XI
1841. FE1 was based on [FC ] and corrected by Chopin.
FE2 Second impression of FE1 (the same firm and number), also
corrected by Chopin.
FE
= FE1 and FE2 .
FED As in the Ballade in G minor, Op. 23.
FES, FEJ as in the Ballade in F, Op. 38.
FESch a copy of FE with annotations by Chopin, from a collection belonging to Chopins pupil, Marie de Scherbatoff
(Houghton Library, New York). Contains fingering, a variant
and a corrected printing error.
EE
First English edition, Wessel & Stapleton (W & S 5299), London I 1842. EE was based on FE2 and was not corrected by
Chopin.

Bars 3 and 39 L.H. In A ([ FC] ,GE ) the note b starting the


second half of the bar is marked with a mordent (which in bar
39 is indicated as
). FE1 contains no ornaments, a fact
which Chopin corrected in FE2 (EE), adding grace-notes
which make it possible to achieve the legato in this phrase
easier than do the mordents. This type of variants is encountered several times in Chopins works e.g. in the Ballade in
G minor, Op. 23, bar 173 or the Mazurka in C minor, Op. 41
No. 4, bars 97-101.

Bar 71 R.H. The main text comes from A ([ FC] ,GE1), the
variant from FE (EE,GE 2). We give priority to the A version since it is not certain whether the a 1 in F E was added in
proof-reading by Chopin himself (it is possible that an error
was committed by the engraver), while the authenticity of the
A version does not give rise to doubts.
Bar 74 L.H. On the last quaver we give the fourth a - d 1 , as in
A ([ FC] ,GE ). In FE (EE) there is an added f . This note has
no natural continuation in the next bar (in contrast to bars
76-77 where it passes to e ), in this way deforming the line
b - a -g-a - b - a -g-f-e of the lower chord notes in bars 73-77.
Probably the engraver of FE mistook this bar for bar 76.
Bar 83 L.H. It does not follow clearly from the script of A
whether the last chord is to have three notes (e - a - e 1 ), as
was interpreted by GE , or four (with c 1 ), as it is recorded in
[ FC] (FEEE).
Bar 86 L.H. FE (EE) overlooks a in the grace-note chord.
Bar 87 We give the octaves on the fourth quaver according to
A ([ FC] ,GE ). The notes a and a 2 were added in FE (EE) .
This is probably an error of the engraver, suggested by the
continuum of previous chords. In A Chopin crossed out the
c 2- a 2- c3 chord in the R.H. and replaced it with an empty octave.
Bar 93 L.H. At the beginning of the bar, A ([ FC] ,GE ) has
only the lower C . In the proofs of FE1 (EE) Chopin added c .

p. 40

Bar 99 The rhythmic record of this bar in the sources is not


clear. A ( [ FC ] FE EE) contains the following script:

Source Commentary

Our attention is drawn to the following discrepancies between


rhythmic values and the graphic arrangement:
in A the a - a 1 octave ascribed to the third quaver of the
bar is located distinctly in the middle of the bar (on the fourth
quaver)
the direction of the stems in the R.H. suggests that the
b-d1 third comprises a second voice, filling the a - a 1 octave.
Basic doubt is produced, therefore, by the moment of striking
the a - a 1 octave on the third or fourth quaver of the bar.
Two solutions come to mind:
1. With the assumption that the proper localisation of particular touches in the bar is described by the rhythmic values, the
a - a 1 octave coincides with the t h i r d quaver of the bar.
This solution is dictated by the following arguments:
Chopins script does not contain an obvious error and was
proof-read by him in A ( on the fifth quaver of the bar was
clearly added later); this fact explains the appearance of the
described discrepancies
Chopin did not alter the rhythm either in [FC] or in FE or in
any of the four pupils copies
it could have been Chopins intention to achieve a rhythmic
and expressive differentiation of bars 99 and 101 (the syncopation in bar 99 would correspond to syncopations in bars
88-94); a similar arrangement of rhythms is encountered in
the Ballade in G minor, Op. 23, bars 167 and 169
despite the fact that the a - a 1 octave was not sustained to
the end of the bar, the direction of the stems clearly indicates
a two-voice treatment of the R.H part. This is the version we
present in the main text.
2. With the assumption that a proper localisation of particular
touches in the bar is determined by the graphic arrangement
of A, it is necessary to correct the values of the rests; this
will lead to the version, presented in the variant, in which the
a - a 1 octave is struck on the f o u r t h quaver of the bar.
This solution is favoured by:
the way of distribution and the rhythmic values of n o t e s
in A; this suggestion is so distinct that GE , based on it,
changed the quaver rests on the second quaver of the bar to
crotchet rests
the rhythmic analogy with bar 101.
Bars 99-100 and 101-102 R.H. The text without ties sustaining melodic sounds comes from A ([ FC]FEEE,
GE1). The o s s i a variants given in the footnote come from
FED (bars 99-100) and FESch (bars 101-102). The variant in
bars 99-100 has two forms corresponding to two possibilities
of interpreting the rhythm in bar 99. In the first the rhythmic
value of the upper a 1 had to be corrected. It must be stressed
that in FES and FEJ the ties have not been added in any of
the two passages, and that Chopin did not add ties in two
places simultaneously in any of the pupils copies. The introduction of two ties in GE 2 was an arbitrary revision, adopted
by the majority of the later collected editions.

(when proof-reading Chopin quite often missed one of the


recurring similar passages).
Bars 109-112 and 150-153 L.H. Chopin could not decide how
to mark the tenor voice distinctive in those bars. In A he originally wrote accents between the notes of the L.H. intervals
but then crossed them out, recognising that the accents-diminuendos between staves were sufficient. However, in print, with
greater spaces in between the staves than in A, this proved
to be unclear for the performers and in pupils copies Chopin
added suitable accents (in FED) or slurs (FES and FEJ).
p. 41

Bar 122 L.H. Certain later collected editions arbitrarily change


the top notes of the chords to g 1 and a 1 .
Bar 125 L.H. In the last chord [FC] (FEEE) overlooks the
note d 1 .
Bar 132 L.H. The main text comes from A ([ FC] ,GE ), and
the variant from FE (EE). The FE version is most probably a simplified correction of a printing error (presumably,
the third quaver was originally completely omitted).
p. 42

Bars 102-103 R.H. The tie joining both g 1 was overlooked in


[ FC] (FEEE).
Bar 107 L.H. The last quaver in the sources is the chord
d-g-b. The analogous bar 56 contains a g-b third, and in bar
60 Chopin crossed out the lower note in a previously written
triad in A. In this situation, it seems highly probable that in
bar 107 Chopin did not correct the triad owing to carelessness

Bar 137 This bar was overlooked in GE .


Bar 138 R.H. The first chord in A has the mistaken value of a
quaver.
Bar 140 R.H. GE 2 arbitrarily ties the a 2 - a

octaves.

Bars 143-144 The bars in A ([ FC] ,GE ) are linked as follows:

Chopin simplified this in the proofs of FE1 (EE).


p. 43

Bar 157 R.H. In A ([ FC] ,GE ) the chord on the third quaver
of the bar has no e 1 . Chopin added this note in the proofs of
FE1 (EE).
Bar 158 R.H. In A (GE ) the note g appears on the fourth
quaver. It is difficult to say whether its absence in [FC] (FE
EE) is an oversight of the copyist or a correction by Chopin.
Bars 160-161 R.H. The tie joining both b was overlooked in
[ FC] (FEEE) and in GE it was reproduced imprecisely.
Bar 162 R.H. It is not clear whether Chopin wished to repeat
or sustain the chord on the fourth quaver of the bar. A (GE1 )
and FE (EE) do not have ties, which, however, appeared in
[ FC] (SC) and have been added by Chopin in FED. Stylistically both versions are possible since this motif appears in
the Ballade in both forms.

Bars 100-101 R.H. Some of the later collected editions arbitrarily join e 1 notes.
Bar 101-102 L.H. The tie joining both e 1 was overlooked in
[ FC] (FEEE). This error was adopted by some of the later
collected editions, with an arbitrary change from e 1 to e 1 at
the beginning of bar 102.

Bar 121 L.H. At the beginning of the bar GE contains a mistaken octave E 1 - E .

p. 44

Bar 173 R.H. Some of the later collected editions arbitrarily


give the triad c 2- e 2 - g 2 as the ninth semiquaver.
Bar 176 In the main text we give the version introduced by
Chopin in the proofs of FE2 (EE). The variant is basically
a version of A ([ FC] ,GE ); only on the third semiquaver in
the R.H. did we remove the note b 1 similarly to the FE2
version (apparently, this simplification of the chord, rather
awkward in high tempo, can be treated as independent of the
remaining changes in this bar).

Source Commentary

Bar 178 R.H. The majority of the later collected editions gives
c 2 instead of the authentic g 2 as the fourth semiquaver.
It seems worthwhile to draw attention to the association of
the span of the R.H. figures with the span of the L.H. chords:
at the end of bar 173 the octave c 3 - c 4 is accompanied by
the chord c 1 - e1 - g 1 - c 2 and in bar 178 the eleventh g 2 - c 4 is
accompanied by the chord g - c 1 - e1 - c 2 .
L.H. FC (FEEE) overlooks the note c 1 in the chord on
the third quaver of the bar.
Bar 179 L.H. A ([ FC] ,GE ) has the F - f octave on the third
quaver. Chopin transferred it an octave higher in the proofs
of FE1 (EE).
p. 45

Bars 195-196 R.H. At the beginning of bar 196 A ([ FC] ,GE )


has a rest in the upper voice. In the proofs of FE1 (EE)
Chopin changed it to a b-d 1 third (tied to the previous one).
Bar 199 R.H. In A the lengthening dots next to the second
chord have been overlooked.
R.H. Next to the lower sound of the last chord in FED Chopin
altered
to . We give this version as o s s i a.
Bar 200 R.H. In the first half of the bar A ([ FC]FE1 ,
GE ) has the following version:

. We accept

the version introduced by Chopin in the proofs of FE2 (EE) .


p. 46

Bar 203 R.H. In the proofs of FE1 (EE) Chopin added a


sustained note c 1 on the fourth quaver of the bar.
Bar 211 R.H. A has the f 1 - f 2 octave on the second quaver;
we also see that Chopin crossed out the inner sound. This
procedure proved to be unclear for readers of A since both
GE and [FC] (FEEE) have the chord f 1- d 2- f 2 at this point.
Bar 213 L.H. A ([ FC]) has a four-note chord on the third
quaver. In it both GE and FE (EE) overlook the note e 1 .
Bar 214 L.H. A ([ FC] ,GE ) has an e 1 - a 1- c 2 chord on the
fourth quaver. In the proofs of FE1 (EE) Chopin removed
the note a 1 .
R.H. In A the fifth quaver is written rather indistinctly and it is
not clear whether a 3 is included in it or not. GE has only c 4,
and [FC] (FEEE) has the a 3- c 4 third.
Bar 215 R.H. The note g 3 in the first chord in A (GE ) is
sustained by a tie but not distinguished as a dotted crotchet.
We correct this imprecision of notation. [FC] (FEEE)
omits the tie joining both g 3.
Bar 216 The main text comes from FE (EE) where it was
introduced by Chopin in his proofs of FE1 (visible traces of
changes in print). The variant is a version of A (GE ). SC
has the same rhythm as FE but the manner of notation testifies to the fact that SaintSans added this rhythm to the A
version i.e. that [FC] had the A version.
Bars 219 and 221 R.H. Some of the later collected editions
arbitrarily add ties sustaining the crotchet g 2 (as in bar 215).
The repetition of g 2 in those bars is connected with a harmonic context in the L.H., different from bar 215.

p. 47

Bars 228-229 R.H. The bars in A ([ FC] ,GE ) are linked as


follows:
of FE1 ( EE) .

10

. Chopin altered this in the proofs

4. Ballade in F minor, Op. 52


S o u r c e s
AI
A 79-bar fragment of an autograph of the first edition of the
Ballade, in the original 6/4 metre (private collection, photocopy in the Chopin Society, Warsaw).
[A1], [A2] the lost first and second of the three autographs which
served as bases for first editions.
A3
Fragment, containing 136 bars, of the chronologically last
autograph/fair-copy (Bodleian Library, Oxford), intended as
the basis for the first German edition. A3 was partially copied
from [A1], and partially from [A2]; it also contains improvements and errors which are absent in all earlier autographs.
[A3 ] Lost end fragment of A 3 .
FE1 First French edition, M. Schlesinger (M.S. 3957), Paris XII
1843. FE1 was based on [A1] and was most probably corrected by Chopin.
FE2 Second impression of FE1 (the same firm and number). Several small changes introduced in FE2 could be the outcome
of its superficial review by Chopin.
FE
= FE1 and FE2 .
FEJ As in the Ballade in F, Op. 38.
EE
First English edition, Wessel & C o ( W & C o 5305), London III
1844. EE was based on [A2] and was not corrected by Chopin; it contains numerous errors.
GE
First German edition, Breitkopf & Hartel

(7001), Leipzig XI
1843. GE reproduces with errors the text of A3 and [A3]; it
was not corrected by Chopin.
E d i t o r i a l P r i n c i p l e s
We have based our text on A3 and, from bar 137, on GE , compared
with FE and EE.
p. 48

Bars 1-2 and 4 GE arbitrarily simplifies Chopins notation,


omitting legato in bar 1, stems and beams distinguishing the
R.H. lower voice in bars 2 and 4, and transferring the L.H.
motif in bars 1-2 to the lower staff. This distorted notation
was repeated in the majority of the later collected editions.
Bar 1-5 R.H. Dynamic markings in parentheses come from FE.
Bar 7 R.H. In the first half of the bar A3 (GE ) mistakenly
repeats the second half of bar 6. FE1 also contains an error:
f 1- c 2 on the fourth semiquaver. The correct text is found in EE
and FE2.
Bars 16 and 30 L.H. The sonority of the second quaver in bar
16 and the fifth quaver in bar 30 gives rise to doubt. In bar 16
all sources have the e 1- f 1- a1 chord but in bar 30 it occurs only
in AI and EE, while FE and A3 (GE ) have the e 1 - a1 interval. In A3 it is clear that in bar 30 Chopin crossed out the f 1
in the originally written chord. Since those bars constitute part
of double-bar phrases, otherwise completely identical, it seems
that it was not the intention of Chopin to differentiate this
detail of the accompaniment. Two hypotheses come to mind:
by crossing out f 1 while looking through A3 Chopin mistook
bar 30 for one of similar bars (21 or 36); therefore, the version concurrent with his intention would be the one which
ignores this crossing-out and which is as a whole and without
errors recorded in [A2] (EE, our main text). In this version,
the presence of the three- or two-note chord is connected
with the preceding harmonic context: in bars 16 and 30 the
first, from several bars appearance of the bass F is accompanied by a three-note chord while bars 21 and 36 (as well as
56 and 150), which end the several-bars fragments based on
F as the pedal point, have a two-note one;
Chopin resigned completely from three-note chords in these
passages, and bar 16 remained uncorrected due to carelessness (when proof-reading Chopin quite often missed one of
the recurring similar passages); we take this possibility into
consideration as a variant.

Source Commentary

p. 50

which is found in AI and A3 (GE ). The FE and EE version


could be an engravers error, possibly committed under the
impact of the previous chord (Chopin wrote notes of seconds
one over the other, and not diagonally next to each other).
This is the reason why we accept d 1 which appears in both
extant manuscripts, and which makes the c 1 - d 1 second
sound in all the chords of the second half of the bar, analogously to the b - c 1 second from the first half.

Bars 43-44 R.H. The main text comes from FE and EE, the
variant from A3 (GE ). Although the absence of ties in A3
could be an oversight on the part of Chopin, both versions
are musically justified:
in a sustention of the chord, the four-bar phrases which start
in the middle of bars 38 and 42 are rhythmically analogous
in a repetition of the chord, there come into being two
pairs of rhythmically corresponding bars bars 38 and 40
and bars 42 and 44.

Bars 72-73 R.H. Some of the later collected editions arbitrarily change the last but one semiquaver in bar 72 from b 1
to b 1 . Others moreover alter the middle notes of chords: in
bar 72 on the eighth and tenth semiquaver from d 1 to e 1 ,
and in bar 73 on the fourth semiquaver from e 1 to f 1 and on
the tenth semiquaver from f 1 to g 1 .

Bar 46 L.H. On the second quaver GE mistakenly has an


additional c 1 .
Bar 48 L.H. The main text comes from AI, EE and A3
(GE ), and the variant from FE. In the main version the
consistent use of chords ensures bars 47-49 a uniform sonority. In the variant, the repetition in bar 48 of intervals from
bar 46 makes it possible for the harmony of the first half of
bar 50, which constitutes a turning point of the harmonic
progression in bars 46-53, to appear more distinctly.

p. 52

Bar 51 L.H. On the second quaver AI, FE and EE have the


chord a - e 1 - g 1 . We accept the version of A3 (GE ) in which
the chords on the second and third quavers have two notes in
common, as in neighbouring figures.

Bar 81 R.H. In FE and EE the melodic note d 1 in the second


half of the bar is not lengthened by a dot or distinguished
from the b - d1 third.

Bar 55 R.H. Some of the later collected editions arbitrarily


change the lower note of the initial chord from e 1 to g 1 .
Bar 56 L.H. In the last chord FE mistakenly omits e 1 . Cf.
bars 16, 21 and analogous ones.
p. 51

Bar 74 L.H. The quaver beginning the second half of the bar,
similarly as the three previous semiquavers, is recorded in the
sources on the upper staff, under the R.H. part. Therefore, by
taking into consideration the allottava sign, it should be read
as f 2 . The shape of figures in the L.H. in the previous two
bars, however, testifies to the fact that Chopin was thinking
of f 1 and made a mistake by transferring the L.H. part to the
lower staff only in the three last semiquavers of the bar.

Bar 83 L.H. The first crotchet in FE is a G-d fifth. We accept


the EE and A3 (GE ) version which stems more naturally from
the previous chord. The absence of d in this version makes it
possible to avoid counter-parallel fifths in relation with the
successive chord. Some of the later collected editions accept
the FE version and arbitrarily add the note d on the last quaver of bar 82.

Bar 58 R.H. Sources differ as regards the sonority of the third


and fourth quaver of the bar. AI has a strictly two-part version:

In [A1] (FE) Chopin supplemented it:

Bar 85-86 R.H. FE has no tie sustaining f 1.

Next, in [A2] (EE):

Bars 87-88 R.H. In A3 (GE ) there are no ties sustaining


the a 1 - b 1- f 2 chord. The repetition of this chord at the end of
the phrasing slur seems to be of little probability and rather
indicates that Chopin overlooked the ties.
p. 53

and in A3 (GE , the version accepted in our edition) Chopin


successively resigned from three-note chords by returning to
a strict two-part texture, but with a line of the lower voice
modified in comparison to previous versions.
1

Bar 59 R.H. FE has c - b


lower voice.

as the first semiquaver of the

Bar 65 R.H. We give the second half of the bar according to


AI and EE. Also A3 (GE ) has a similar version, which differs only because the a 1 note in the chord is not lengthened
to a dotted crotchet. FE contains the following version:
. The lower note of the chord (g 1 )
could be an error made by the engraver. The authenticity of
note a 1 repeated on the last but one semiquaver does not
give rise to doubts but its absence in EE and A3 probably
signifies that Chopin ultimately resigned from it.
Bar 68 R.H. GE mistakenly ties notes a 1 in the two first
quavers.
Bar 72 R.H. In the chord on the second semiquaver of the
second half of the bar FE and EE have e 1 instead of d 1

Bar 92 R.H. In the middle of the bar FE does not lengthen note
d 2 with a dot and on the last quaver it has a sixth d 1 - b 1 . We
accept the EE and A3 (GE ) version (in A3 Chopin crossed
out b 1 and added a dot lengthening the sonority of d 2 ).
Bars 94-95 Some of the later collected editions arbitrarily
add ties joining chords over the bar-line.
Bar 99 L.H. At the beginning of the second half of the bar FE
and EE have the chord B -f-b . We accept the A3 (GE )
version.
Bars 99-100 R.H. GE overlooks the tie sustaining d 1 .
Bar 104 L.H. In the main text we give the A3 (GE ) version
in which the melody in the octaves stems in the most natural
way from the five-semiquaver transition in bar 103. The version given in the footnote comes from FE and EE.
Bars 108-109 and 110-111 R.H. In A3 (GE ) the sixth d 2 - b 2
in bars 108-109 is sustained by ties. In analogous bars 110-111 A3 discloses crossed out ties sustaining c 2 - a 2 . FE and
EE do not have ties in any of those passages. In the majority
of the later collected editions the sixths are tied arbitrarily in
both pairs of bars. The A3 version, with a differentiation of
those passages, characteristic for Chopin, is most probably
the final one.

11

Source Commentary

p. 54

Bar 154 L.H. The last semiquaver in GE is most probably the


mistaken c 2 . This sort of an error was committed at times by
Chopin and particularly often by the engravers of GE (e.g. in
bar 159, where the last semiquaver is c 2 instead of a 1 ).

Bar 113 R.H. The first half of the bar in EE sounds as follows:
and in A3 (GE ):
These versions can testify to the fact that Chopin wavered in
choosing the best place for interrupting the continuous progression of the sixths; they could be also the consequence of
the mistakes made by Chopin and the engraver of EE. We
accept the FE version since it is:
musically the smoothest and pianistically the most convenient
confirmed by a concurrent version of the analogous bar 115.
Bar 118 L.H. The main text comes from FE and A3 (GE ) ,
the variant from EE. An interval on top of the progression
of chords is one of the devices used by Chopin (e.g. the
Ballade in G minor, Op. 23, bar 207), while the three-note
chord alleviates somewhat the effect of the simultaneously
struck doubled third g 2- g 3 .

p. 57

Bars 164 and 165 R.H. Prior to the fifth semiquaver EE in


both bars and GE in bar 165 add . This sort of supplementation is one of the most frequent arbitrary revisions of those
editions (cf. commentary to the Ballade in G minor, Op. 23,
bars 45 and 47). In both cases FE has flats.
p. 58

p. 55

Bar 176 L.H. The main text comes from FE and GE , and the
variant from EE.
p. 59

p. 60

p. 61

p. 62

Bar 150 R.H. The main text comes from FE and EE, the variant from GE . The absence of [A3] makes it impossible to
say whether the sustaining of e 2 is an authentic variant or an
error of GE .
Bar 153 R.H. In the first half of the bar FE has the following
rhythm:

12

Bar 226 R.H. In the last chord there is no c 2 in FE.


p. 63

Bar 144 L.H. On the fifth and sixth quaver FE1 has no g or
f 1 . In FE2 f 1 is supplemented, possibly upon the basis of
[A1]. We give the EE and GE version.

Bar 217 L.H. The main text comes from EE and GE . In the
variant we give a pianistically slightly more difficult version of
FE, which offers fuller harmony.

Bars 223 and 225 R.H. Some of the later collected editions
arbitrarily change f 1 to e 1 on the first semiquaver.

Bar 134 R.H. Note a 1 on the sixth semiquaver in FE has the


value of a semiquaver; in EE this sound is written as a quaver tied with an additional dotted crotchet.

Bars 198-202 The crescendo sign and


come from FE,
where Chopin probably added them in the proofs.

Bar 221 L.H. The GE version given in the footnote may be


mistaken: Chopin frequently placed notes under the first line
much lower, which in this case could have inclined the engraver to read F as D .

Bar 130 R.H. On the fourth semiquaver FE has no g 1 .

p. 56

Bar 190 L.H. The main text comes from EE and GE , and the
variant from FE. From the point of view of sonority the wider
figuration of the FE version corresponds to the development
of the theme as a whole (from bar 169). In EE and GE Chopin
resigned from it in favour of a passage remaining within the
framework designated by the directly preceding figures. Cf. a
similarly justified variant at the end of the Ballade in F, Op. 38.

Bar 200 R.H. Some of the later collected editions arbitrarily


add c 2 to the first chord.

Bars 127-128 L.H. GE overlooks the tie linking A with G .

Bar 136 N. B. A3 ends with this bar.

Bar 184 L.H. The main text comes from GE , and the variant
from FE and EE.
Bar 185 L.H. As the fourth and sixth semiquaver FE has the
most probably mistaken d 1 and B .

Bars 125-126 L.H. A3 (GE ) overlooks the tie sustaining b .


Bar 127 L.H. The main text comes from EE, the variant
from FE and A3 (GE ). We give priority to the EE version,
unquestioned in this place, since we cannot exclude the possibility that the [A1] (FE) version, copied in A3, is only an
earlier notation, deformed by the overlooking of a tie sustaining A on the third eighth.

Bar 167 R.H. The main text comes from FE and GE , the
variant from EE.
Bar 174 R.H. At the beginning of the second half of the bar
FE has the chord d 1 - c 2 - d 2 . We give the final version of EE
and GE .

Bar 123 R.H. A3 (GE ) overlooks the tie sustaining a 1 .


Bar 124 R.H. The oversights of accidentals occurring in some
sources in the second half of the bar the absence of next
to c 2 in F E and A3 (GE ) and the absence of
next to f 2 in
A3 (GE ) inclined the editors of the majority of the later
collected editions to change those notes to c 2 and f 2 . It
should be stressed that f 2 on the last but one semiquaver does
not give rise to doubts, neither as regards the sources (the
flat is absent in all cases, and two sources contain naturals)
nor musical (cf. a similar device which anticipates the major
key in the Sonata in B minor, Op. 35, first movement, bar
168 and a similar juxtaposition of f and f in adjoining octaves
in this Ballade, bar 62, and in the Waltz in A , Op. 34 No. 1 ,
bars 251-252), while c 2 in the ninth semiquaver is justified by
tonal relations (despite the false modulations to D minor in
bars 121-124, A major remains the basic key from bar 115
on) and gives a smoother progression of the top semiquavers
in each pair (f 1- c 2- f 2 ) .

Bars 156-157 R.H. GE has no tie sustaining a 2 .

Bar 227 L.H. Instead of a semiquaver triplet, EE has a F 1 - F


quaver. This version, analogous to the beginning of bar 225,
was probably replaced by Chopin with the FE and GE version, which does not interrupt the sequence of semiquavers
and gives a logical beginning to the figures in the R.H.
L.H. In FE and GE the second semiquaver in the second half
of the bar is G. This is probably an original or mistaken version of [A1] (FE), copied owing to carelessness in [A3 ]
(GE ). We give the EE version, which does not give rise to
doubts and is analogous to bars 228-230.
Jan Ekier
Pawe Kamiski